If It Flies ... - 2010/06

HI GANG. It’s time to continue our feet-onthe-
ground test-pilot school. Test-flying our
aircraft is a practice that separates us
aeromodelers from all but a relative handful
of full-scale pilots: the professional test pilots
and the amateurs, otherwise known as homebuilders.
We aeromodelers actually share a great
deal with the home-builders, including a fair
number of people. The overlap between us
and full-scale pilots is heavily populated by
home-builders, or those who belong to the
EAA (Experimental Aircraft Association).
That group has experience and
information to share with both groups, and
many of us would do well to “debrief” these
pilots over a friendly cup of coffee. Flying
stories are much the same, whether models or
full-scale, and descriptively hand-flying
serves as a common language.
In some cases, the full-scale homebuilders
become modelers to gain access to
the poor man’s wind tunnel, which is also
known as the sky. Testing the stability of an
airplane near the aft CG limit, spin-recovery
testing, and even dive and flutter testing is
much less risky if you have done the tests on a
scaled-down prototype with your feet safely
on the ground.
Full-scale flight-testing is a methodical,
step-by-step process, because you need to
sneak up on an aircraft’s bad behavior without
taking reckless, life-endangering risks—even
though you may not know exactly where or
how the airplane’s nastiness will begin.
The aircraft’s safe-flight envelope is
pushed open wider and wider, and when the
limits begin to show themselves, the pilot’s
manual reflects those with some safety margin
built in. Flying is
generally safe as a
result.
Before a fullscale
student pilot
climbs into a Cessna
152 or similar
trainer, the instructor
will spend at least a
little time discussing
the permissible
weight and balance
calculation and chart
in the pilot’s manual.
When pilots fail to
adhere to this
procedure, we see
pictures of the
wreckage on TV and
read about it in more depth later.
For some reason, the spectacular crashes
often seem to involve famous passengers
who refuse to travel without entourages and
heavy baggage …
The professional test pilot’s superior
piloting skills are called for when the test
airplane’s behavior deteriorates abruptly and
unexpectedly: when slow or incorrect first
responses might result in a smoking airplaneshaped
hole in the ground.
With rare exceptions, methodical testing
almost turns the test pilot’s job into another
day at the office. Of course, the stakes are
nowhere nearly as high for us aeromodelers;
our feet are safely on the ground. So why
approach this test-flight stuff so
scientifically?
I don’t know about you, but I hate to
pound my investment in both time and
money into little pieces, especially when I
designed the thing. No matter how
disposable you might consider your models
to be, they do create safety hazards when
they are not fully under control.
I’m not looking to step on MA safety
columnist Dave Gee’s toes, but through the
years I have seen many airplanes flown
repeatedly in front of crowds, despite the fact
that they were not properly shaken down in
early flight-testing. An instance that is vivid
in my mind was a high-wing competition
Scale airplane that was flown for years in
contests near where I live.
It veered uncontrollably to the left after
each takeoff and overflew the pits as a result.
A year later, I watched the same model, and
pilot, do the same thing in the same place on
the same runway. I changed where I parked
the car on year three.
Feet-on-the-ground test-pilot school
June 2010 79
Dean Pappas | DeanF3AF2B@If It Flies ... pappasfamily.net
Also included in this column:
• The tuck-under CG condition
• Fine-tuning pitch stability
• Flutter testing
Flying wings such as this Rainbow and conventional-layout models with highly cambered wing
airfoils and short tails are subject to high-speed pitch instability known as “tuck-under.” The
airplane gains speed in a dive and pitches nose-down, eventually stabilizing in a steep inverted
dive. Full-scale test pilots have died from this. The cure is to move the CG forward until the
instability is manageable or gone. Gordy Stahl photo.
Below: The first time Dean heard of high-speed
tuck-under in a model was in the July 1976
Flying Models magazine, in Dick Sarpolus’s
construction feature on the LARS (Low Aspect
Ratio Sailplane). A good textbook later
described what had happened, but when the
author encountered it some 18 years later, the
understanding of what was going on likely saved
the airplane. The LARS has a short tail
compared to similar 2-meter sailplanes, because
aerodynamic tail length is measured in average
wing chords. Sarpolus photo.
06sig3.QXD_00MSTRPG.QXD 4/23/10 9:13 AM Page 79
There is no need to poke at this pilot’s
flying skills, because the airplane obviously
needed changes that obviously would have
enhanced my safety and his competition flying
scores.
It has to be tough to overcome getting a
zero score on Takeoff on each flight. The
changes would have been minor, but it
requires adopting the mind-set that you will
make a small change, hopefully based on the
kinds of things we discussed here in the last
few months, and then go fly the model with an
eye toward critically evaluating what you have
accomplished.
If need be, get a flying buddy to watch
carefully. Two sets of eyes might be better
than one, especially when that first set is busy
flying.
The Dreaded Tuck-Under: In the past few
columns I’ve written about engine side thrust
and downthrust, roll control, adverse yaw, and
the evils of cross-trimming. In each case I
covered the flight test. Sometimes it is simple,
and other times it takes patience and a few
tries. Maybe there is something to those testpilot
skills after all.
Awhile back I wrote about unbalanced
wing weight. If I remember correctly, I
covered how an airplane tends to wander off in
the direction of the heavy wingtip at the most
critical stage of landing. Now let’s complete
the test-pilot course by reviewing the effects of
the fore and aft balance point, or CG.
Depending on the type of model and the
type of flying you do, different tests are called
for when it comes to investigating the effects
of CG position. I’ll start with the most
dramatic one I can think of, only because this
one is fun.
Flying wings or tail-less aircraft have
special stability issues that are shared only by
airplanes with tails that have highly cambered
wings and short tail moments. When these
models are balanced for optimum
performance, they become sensitive to high
airspeed in a special way.
Normally, assuming that an aircraft is
nose-heavy enough for good stability, its pitch
trim becomes airspeed-sensitive in the positive
sense: the faster it goes, the more it tends to
climb. The slower its airspeed, the more it
tends to drop its nose. If you think about it,
that’s a nice, stable state of affairs.
When a flying wing has a forward CG, it
will be stable but it will also waste much of its
lift fighting the nose-down pitching tendency
that the airfoil has. For this reason, flying
wings either have TE reflex or elevons trailed
up.
If the flying wing’s CG is far enough aft to
achieve close-to-optimum performance, the
amount of reflex or up-elevon is reduced, as is
the drag this creates. The model also develops
a critical airspeed above which this normal
airspeed/pitch relationship turns backward.
With increasing airspeed, the airplane
tends to nose down. The resulting dive leads to
even higher airspeeds. Left unchecked, the
aircraft will restabilize in a high-speed,
inverted 45° dive. That usually ends badly.
Sometimes the aircraft responds to upelevator
and a reduction in power (if not a
glider), and sometimes it does not. The
difference is often in how quickly the
correction is made. What to do?
The flight test itself is simple. Do the test at
full throttle if you have an engine or motor,
because when the tuck-under shows up you
can simultaneously pull the power back and
pull up-elevator.
Start the test in full-power level flight,
and on the next pass use a shallow (5°)
dive. Make the next dive a little steeper,
and the next should be even steeper, until
you are convinced that you will never get
the airplane moving this fast and the
problem still hasn’t occurred or until you
are flying fast enough to concern yourself
with control-surface flutter.
In that case, the course of action is the
same; pull off the power quickly and pull up
the nose to kill the airspeed. Conduct these
tests high enough to give yourself some
recovery room, and for heaven’s sake perform
them pointed away from any people.
Those methodical steps I wrote about
before are important now. Start noticeably
nose-heavy and move the CG aft a tiny bit
each time the flight test is successful. Small
CG movements assure that the problem won’t
be severe when it rears its ugly head, and
human reflexes will be fast enough to stop the
tuck-under before it fully develops. Then go
back to the last good CG position.
Maybe 15 years ago, a flying buddy of
mine contracted with a local home-builder/
designer to produce a stability test model of a
flying-wing ultralight he planned to build and
for which he hoped to sell plans. Repeated
tests showed that the aircraft had a pronounced
tuck-under when balanced at the designer’s
calculated CG.
If the full-scale pilot’s arms were not
strong enough to counteract the nose-down
pitch, this sort of thing could have killed him.
As it was, the roughly 2,000-square-inch test
model almost overpowered the Giant Scale
servos, and I nearly lost the airplane the first
time we did this test.
When we found the real safe aft CG
location, the designer got disgusted with the
projected performance and abandoned the
project. He is still alive, though.
Fine-Tuning Pitch Stability: Most trainers,
especially those with flat-bottomed airfoils, are
intended to be nice and stable; that is, they
tend to nose up when airspeed increases and
drop the nose when airspeed decreases from
the speed at which the model was trimmed. As
I mentioned, that is a good thing.
Too much of a good thing means that the
airplane tends to climb too steeply at full
power and that the glide ends up too steep for
comfort, all with the same elevator trim
setting. Curiously enough, this is what a
Cessna 152 will do when trimmed for level
cruise flight. The full-scale pilots constantly
retrim the elevator based on what they are
doing.
If the pitch stability is marginal, the fullpower
climb might require that you add upelevator,
which is no big deal, but the glide
angle might be extremely shallow with low
airspeed and poor roll control authority.
(Astute readers might remember a similar
discussion from roughly a year ago, but it is
repeated here for the sake of completeness.)
The flight test looks like this. Trim the
elevator for level flight at the same throttle
setting that you normally intend to use for
cruising flight. The following checks are done
with that same elevator trim setting.
From level flight and with your hands off
of the stick (unless a problem develops), add
full throttle and watch what develops. Is the
climb too steep? Does the model pitch up,
stall, and drop the nose? On the other hand,
does it simply fly faster without climbing
much? Take note of what the aircraft does and
then move on to the power-off test.
Gain a comfortable bit of altitude and,
while in hands-off level flight at the same
cruise power and elevator trim setting, reduce
the throttle to idle and wait to see what
happens. The model will eventually decelerate
and settle into a glide.
Is the glide slope and speed acceptable? Is
it too fast and steep or too shallow with poor
directional control because of low airspeed?
If you decide to push the CG aft to reduce
the difference between climb and glide trim or
move it forward to steepen the glide, you need
to retrim the elevator for level cruise flight
before retesting to see if you like the result.
Sport and aerobatic models with
semisymmetrical and symmetrical airfoils are
intended for maneuvering. In general, they are
flown with the CG far enough aft that only
slight (but some) down-elevator is required for
level inverted flight and just far enough
forward that the airplanes “groove” well and
that their elevators are not jumpy near neutral.
The checks for both are straightforward,
and both are a matter of feel rather than having
a nice, neat test. Symmetric-airfoil aircraft are
not subject to the tuck-under problem at high
speed, but under power-off conditions they
suffer from slow, shallow glides and poor
control when tail-heavy.
Flutter Testing (aka Flirting With Disaster):
The best way to avoid control-surface flutter is
to build light, stiff control surfaces and to
connect them with short, stiff linkages to
servos that are not undersized. However, many
airplanes suffer from this condition.
It is often discovered only after the aircraft
has aged and the servo gears wear a bit and the
covering loosens slightly. Control-surface
flutter typically happens in a high-speed dive.
If the pilot is fortunate enough to immediately
hear the angry buzzing sound and promptly
take action, he or she might get to keep the
airplane.
Sometimes it’s too late when you hear the
“braaaap” of a flapping control surface. For
that reason, pushing an airplane faster in
increasingly steeper dives at full power is a
dumb idea.
However, if you anticipate flying an
aircraft very fast on a regular basis, it is a good
idea to work up to maximum airspeed in steps,
looking and listening carefully for the tiniest
hint of flutter, whether it be the sound or a
sudden tendency for it to wander off to one
side. This often happens during aileron flutter,
which is the most common variety.
Do these tests when the flightline is quiet
so you can hear any flutter, and keep the model
pointed away from people. If flutter starts, the
course of action is the same as I have
described.
Pull off the power quickly, and quickly but
smoothly pull the nose up to kill the airspeed.
keep the airspeed moderate afterward, and
land immediately; the strangest things loosen
in an airplane that has undergone flutter.
Check all hinges, linkages, servo mounts, and
even engine mount.
That wraps up the feet-on-the-ground testpilot
school. In the next column I’ll write about
something different. Until then, have fun and
do take care of yourself. MA